CN102104433A - Symbol detection method and equipment - Google Patents

Abstract

The invention provides symbol detection equipment at the receiving terminal of a mobile communication system. The symbol detection equipment comprises a receiving device, a channel estimation device and a balance, hard decision and demodulation device, wherein the receiving device is used for receiving and extracting a subframe in data from an emitting end; the channel estimation device is used for carrying out channel estimation on a pilot symbol of the received subframe so as to acquire the channel response of the pilot symbol, and carrying out channel estimation on a data symbol by utilizing a result obtained through balance, hard decision and demodulation, of a data symbol adjacent to the pilot symbol so as to acquire the channel response of the data symbol, wherein the result is obtained through balancing, hard decision and demodulation is acquired by a balance, hard decision and demodulation device; and the balance, hard decision and demodulation device is used for executing balance, hard decision and demodulation to the data symbol by utilizing the channel response of the pilot symbol, and executing balance, hard decision and demodulation on another data symbol adjacent to the data symbol by utilizing the channel response of the data symbol, acquired from the channel estimation device.

Description

Symbol detection method and symbol detection apparatus

Technical field

The present invention relates to moving communicating field, more specifically, proposed a kind of symbol detection method and symbol detection apparatus of receiving terminal place of mobile communication system, can improve the receptivity under high speed and high SNR environment.

Background technology

Under high velocity environment, channel response is along with the time changes highly significant, thereby how coming accurately that the channel response of data estimator symbol has become hot technology according to channel estimation results to frequency pilot sign, this is because channel response has remarkable influence to final detection performance.

In order to address this problem, the present invention proposes a kind of combined channel estimates and symbol detection feedback algorithm, be used for the detection performance of LTE (Long Term Evolution) the up link receive diversity under high speed and high s/n ratio (SNR) environment, and have acceptable implementation complexity.

Summary of the invention

For the defective that overcomes prior art has proposed the present invention.Therefore, the objective of the invention is to propose a kind of symbol detection method and symbol detection apparatus of receiving terminal place of mobile communication system, can improve at a high speed and the receptivity under the high SNR environment.

To achieve these goals, according to the present invention, proposed a kind of symbol detection method of receiving terminal place of mobile communication system, having comprised: the frequency pilot sign to the subframe that receives carries out channel estimating, to obtain the channel response of frequency pilot sign; Utilize the channel response pair data symbol adjacent of pilot signal to carry out equilibrium, hard decision and demodulation with frequency pilot sign; Utilize the result after equilibrium, hard decision and the demodulation of described data symbol that described data symbol is carried out channel estimating, to obtain the channel response of described data symbol; And the channel response that utilizes resulting described data symbol, another data symbol adjacent with data symbol carried out equilibrium, hard decision and demodulation.

Preferably, described method also comprises: utilize the result after equilibrium, hard decision and the demodulation of described another data symbol that described another data symbol is carried out channel estimating, to obtain the channel response of described another data symbol; And the channel response that utilizes resulting described another data symbol, an again data symbol adjacent with another data symbol carried out equilibrium, hard decision and demodulation, and the like.

Preferably, described method also comprises: the channel response of described frequency pilot sign and data symbol is carried out filtering, to remove residual noise.

Preferably, described frequency pilot sign and described data symbol are the orthogonal frequency division multiplex OFDM symbols that is carried on a plurality of subcarriers.

Preferably, described mobile communication system is applied to the up link receive diversity of Long Term Evolution LTE system.

To achieve these goals, according to the present invention, also proposed a kind of symbol detection apparatus of receiving terminal place of mobile communication system, having comprised: receiving system receives and extracts from the subframe in the data of transmitting terminal; Channel estimating apparatus, frequency pilot sign to the subframe that receives carries out channel estimating, to obtain the channel response of frequency pilot sign, and utilize the result after equilibrium, hard decision and the demodulation of the data symbol adjacent that obtains from balanced, hard decision and demodulating equipment that described data symbol is carried out channel estimating with frequency pilot sign, to obtain the channel response of described data symbol; And equilibrium, hard decision and demodulating equipment, utilize the channel response of pilot signal that described data symbol is carried out equilibrium, hard decision and demodulation; And utilization is carried out equilibrium, hard decision and demodulation from the channel response of the resulting described data symbol of channel estimating apparatus to another data symbol adjacent with data symbol.

Description of drawings

Below in conjunction with the detailed description of preferred embodiment of accompanying drawing to being adopted, above-mentioned purpose of the present invention, advantage and feature will become apparent by reference, wherein:

Fig. 1 shows the schematic diagram of LTE uplink frame structure;

Fig. 2 shows the schematic diagram of the linear interpolation algorithm of prior art;

Fig. 3 shows the schematic diagram of prior art (least mean-square error) MMSE filtering algorithm;

Fig. 4 is the flow chart of symbol detection method according to an embodiment of the invention; And

Fig. 5 is the block diagram of symbol detection apparatus according to an embodiment of the invention.

Embodiment

Fig. 1 shows the frame structure of LTE up link.1 subframe continues 1ms, and is made up of 14 OFDM symbols.Symbol 4 and 11 is used for carrying pilot tone, is called as frequency pilot sign, and other symbols are used for carrying user data, thereby is called as data symbol.The number of subcarriers of each shown in the figure (OFDM) OFDM symbol determines by system bandwidth, and in the LTE of 10MHz system N=1024.

In the LTE of reality system, there are two kinds of general time domain interpolation algorithms: (1) linear interpolation; And (least mean-square error) MMSE filtering of (2) time domain.

Linear interpolation algorithm supposition channel response (amplitude and phase place) as shown in Figure 2 is to be constant or linear change in the given duration (for example 1 TTI (Transmission Time Interval) or 1 time slot).Therefore, according to channel estimation results, can estimate channel response effectively according to above hypothesis to the data symbol to two frequency pilot signs.This constant channel hypothesis is reasonably under the time-varying characteristics of wireless channel and unconspicuous low speed situation of movement, and under the high-speed mobile situation, and is because the time-varying characteristics of channel response become all the more obvious, therefore should hypothesis no longer reasonable.In classical Doppler frequency spectrum, the correlation function of time varying channel is followed Bessel function, and this function has tangible nonlinear characteristic, so linear hypothesis be not suitable under the high-speed mobile situation yet.

The correlation function of MMSE filtering algorithm supposition time varying channel is as shown in Figure 3 followed the Bessel function with two parameters, these two parameters are speed and SNR, and according to the accurate estimation to speed and SNR, this algorithm can obtain than the better channel estimating of linear interpolation algorithm.H among Fig. 3 ₁(i) and H ₂(i) channel estimating of 2 pilot tones on i subcarrier of expression.Yet the performance height of this MMSE filtering algorithm depends on priori channel model and velocity estimation, thereby owing to following 3 performances that reason can't obtain in real system:

(1) given priori channel model can't cover all communication environments;

(2) in real system, be difficult to obtain robust and accurate velocity estimation;

(3) time domain is relevant under high-speed case become a little less than.

Obviously, these two kinds of algorithms all can't provide gratifying solution under high velocity environment.

According to the present invention, the combined channel that has proposed a kind of LTE of being applied to up link receive diversity is estimated and the symbol detection feedback algorithm.This basic idea is: balanced data is carried out hard decision, according to receiving data and hard decision result the data symbol is carried out channel estimating then, and channel estimation results is used for adjacent data OFDM symbol is carried out equilibrium.Usually, compare with non-adjacent OFDM symbol, the channel between the adjacent OFDM symbol more may keep constant or linear change.

The handling process of algorithm of the present invention is as described below, and wherein supposition is in the LTE system of 10MHz.Suppose in up link, to have unique user, and transmitting of unique user can be expressed as:

$S[n,k]=\left\{\begin{array}{ccc}P[n,k]& n=\mathrm{4,11},& k\∈\left[\mathrm{0,1023}\right]\\ X[n,k]& n=\mathrm{1,2,3,5,6,7,8,9,10,12,13,14},& k\∈\left[\mathrm{0,1023}\right]\end{array}\right.---\left(1\right)$

P[n wherein, k] and X[n, k] represent frequency pilot sign and data symbol respectively, and n represents n OFDM symbol.K represents k subcarrier.Received signal can be expressed as in frequency domain:

R[n，k]＝S[n，k]·H[n，k]+W[n，k] (2)

H[n wherein, k] be the channel response in the frequency domain, and W[n, k] be the FFT result of AWGN (additive white Gaussian noise).At receiver side, least square (LS) channel estimating of frequency pilot sign be can be expressed as:

$H_{\mathrm{LS}}[i,j]=\frac{R[i,j]}{P[i,j]}=H[i,j]+\frac{W[i,j]}{P[i,j]},i=\mathrm{4,11};j\∈\left[\mathrm{0,1023}\right]---\left(3\right)$

As the initial input of entire process, should be accurate as far as possible to the channel estimating of frequency pilot sign.In order to remove residual noise, should in frequency domain or time domain, carry out filtering to the channel estimation results in the equation (3).According to accurate channel estimating, can realize algorithm of the present invention according to following steps as shown in Figure 4 to frequency pilot sign.

The first step is carried out channel estimating to frequency pilot sign 4 and 11 (frequency pilot sign), and this result (channel response of frequency pilot sign) is used as initial input:

$\hat{H}[i,k]=\mathrm{filter}\left(\frac{R[i,k]}{P[i,k]}\right),i=\mathrm{4,11}---\left(4\right)$

Function " filter " expression time-domain filtering wherein below also is like this

In second step, utilize the channel estimating of frequency pilot sign 4 carried out data symbol 3 adjacent with frequency pilot sign 4 and 5 equilibrium, hard decision and demodulation:

$\hat{S}[i,k]=\frac{1}{{\left|{\hat{H}}_{00}\right[4,k\left]\right|}^2+{\left|{\hat{H}}_{10}\right[4,k\left]\right|}^2+{\mathrm{\σ}}^2}\left[\begin{array}{cc}{\hat{H}}_{00}{[4,k]}^{*}& {\hat{H}}_{10}{[4,k]}^{*}\end{array}\right]\·R^H[i,k],i=\mathrm{3,5}---\left(5\right)$

$\hat{C}[i,k]=\mathrm{de}\mathrm{mod}\mathrm{ulate}\left(\hat{S}\right[i,k\left]\right),i=\mathrm{3,5}---\left(6\right)$

Wherein, the mediation hard decision is separated in function " demodulate " expression, below also is like this.

In the 3rd step, utilize the channel estimating of frequency pilot sign 11 carried out data symbol 10 adjacent with frequency pilot sign 11 and 12 equilibrium, hard decision and demodulation:

$\hat{S}[i,k]=\frac{1}{{\left|{\hat{H}}_{00}\right[11,k\left]\right|}^2+{\left|{\hat{H}}_{10}\right[11,k\left]\right|}^2+{\mathrm{\σ}}^2}\left[\begin{array}{cc}{\hat{H}}_{00}{[11,k]}^{*}& {\hat{H}}_{10}{[11,k]}^{*}\end{array}\right]\·R^H[i,k],i=\mathrm{10,12}---\left(7\right)$

$\hat{C}[i,k]=\mathrm{de}\mathrm{mod}\mathrm{ulate}\left(\hat{S}\right[i,k\left]\right),i=\mathrm{10,12}---\left(8\right)$

The 4th step, equilibrium, hard decision and the demodulation result of data symbol 3,5,10 and 12 are modulated, then modulation result is fed back to channel estimation module:

$\hat{P}[i,k]=\mathrm{mod}\mathrm{ulate}\left(\hat{C}\right[i,k\left]\right),i=\mathrm{3,5,10,12}---\left(9\right)$

Wherein, function " modulate " expression modulates the hard decision result, below also is like this.

The 5th step, data symbol 3,5,10 and 12 is carried out channel estimating, to obtain the channel response of data symbol 3,5,10 and 12:

$\hat{H}[i,k]=\mathrm{filter}\left(\frac{R[i,k]}{\hat{P}[i,k]}\right),i=\mathrm{3,5,10,12}---\left(10\right)$

The 6th step, utilize to the channel estimation results of data symbol 3 and 5 carry out to data symbol 3 and 5 adjacent data symbols 2 and 6 equilibrium, hard decision and demodulation, be similar to the second above step.Then, equilibrium, hard decision and the demodulation result of data symbol 2 and 6 are modulated, with modulation result

[2, k] and

[6, k] feed back to channel estimation module, to obtain channel response to data symbol 2 and 6:

[2, k] and

[6, k] are similar to the 5th above step.

The 7th step, utilization to the channel estimation results of data symbol 10 and 12 carry out to data symbol 10 and 12 adjacent data symbols 9 and 13 equilibrium, hard decision and demodulation, be similar to above the 3rd step, then, equilibrium, hard decision and demodulation result to data symbol 9 and 13 are modulated, with modulation result

[9, k] and

[13, k] feed back to channel estimation module, to obtain the channel response to data symbol 9 and 13

[9, k] and

[13, k] are similar to the 5th above step.

The 8th step, utilize the channel response of data symbol 2 and 6 carry out to data symbol 2 and 6 adjacent data symbols 1 and 7 equilibrium, hard decision and demodulation, be similar to the second above step.

The 9th step, utilize the channel response of data symbol 9 and 13 carry out to data symbol 9 and 13 adjacent data symbols 8 and 14 equilibrium, hard decision and demodulation, be similar to the 3rd above step.

Fig. 5 is the block diagram of symbol detection apparatus according to an embodiment of the invention.

As shown in Figure 5, symbol detection apparatus according to the present invention comprises

According to algorithm of the present invention can at a high speed and provide under the high SNR environment than existing method better with the receptivity of robust more.The implementation complexity of this algorithm and processing latency are acceptable for real system, thereby have higher practicality.

With respect to linear interpolation algorithm, can provide better ability of following the tracks of nonlinear time-varying channel under high speed and the high SNR environment according to algorithm of the present invention, thereby have better detection performance than linear interpolation algorithm.

With respect to the MMSE filtering algorithm, because it is independently and for the precision of velocity estimation and insensitive on given channel model, thereby according to algorithm of the present invention robust more.This complexity that also shows algorithm of the present invention will because if the MMSE filtering algorithm goes for more performance, then need more resources to obtain robust and precise channels parameter less than the MMSE filtering algorithm.Therefore, algorithm according to the present invention has better actual performance than MMSE filtering algorithm.

Although below show the present invention in conjunction with the preferred embodiments of the present invention, one skilled in the art will appreciate that under the situation that does not break away from the spirit and scope of the present invention, can carry out various modifications, replacement and change to the present invention.Therefore, the present invention should not limited by the foregoing description, and should be limited by claims and equivalent thereof.

Claims (9)

1. the symbol detection method at the receiving terminal place of a mobile communication system comprises:

Frequency pilot sign to the subframe that receives carries out channel estimating, to obtain the channel response of frequency pilot sign;

Utilize the channel response pair data symbol adjacent of pilot signal to carry out equilibrium, hard decision and demodulation with frequency pilot sign;

Utilize the result after equilibrium, hard decision and the demodulation of described data symbol that described data symbol is carried out channel estimating, to obtain the channel response of described data symbol; And

Utilize the channel response of resulting described data symbol, another data symbol adjacent with data symbol carried out equilibrium, hard decision and demodulation.

2. method according to claim 1 also comprises:

Utilize the result after equilibrium, hard decision and the demodulation of described another data symbol that described another data symbol is carried out channel estimating, to obtain the channel response of described another data symbol; And

Utilize the channel response of resulting described another data symbol, an again data symbol adjacent with another data symbol carried out equilibrium, hard decision and demodulation, and the like.

3. method according to claim 1 also comprises:

The channel response of described frequency pilot sign and data symbol is carried out filtering, to remove residual noise.

4. method according to claim 1, wherein said frequency pilot sign and described data symbol are the orthogonal frequency division multiplex OFDM symbols that is carried on a plurality of subcarriers.

5. method according to claim 1, wherein said mobile communication system are applied to the up link receive diversity of Long Term Evolution LTE system.

6. the symbol detection apparatus at the receiving terminal place of a mobile communication system comprises:

Receiving system receives and extracts from the subframe in the data of transmitting terminal;

Channel estimating apparatus, frequency pilot sign to the subframe that receives carries out channel estimating, to obtain the channel response of frequency pilot sign, and utilize the result after equilibrium, hard decision and the demodulation of the data symbol adjacent that obtains from balanced, hard decision and demodulating equipment that described data symbol is carried out channel estimating with frequency pilot sign, to obtain the channel response of described data symbol; And

Equilibrium, hard decision and demodulating equipment utilize the channel response of pilot signal that described data symbol is carried out equilibrium, hard decision and demodulation; And utilization is carried out equilibrium, hard decision and demodulation from the channel response of the resulting described data symbol of channel estimating apparatus to another data symbol adjacent with data symbol.

7. symbol detection apparatus according to claim 6, wherein

Result after equilibrium, hard decision and the demodulation of described another data symbol that described channel estimating apparatus utilization obtains from balanced, hard decision and demodulating equipment carries out channel estimating to described another data symbol, to obtain the channel response of described another data symbol; And

Described equilibrium, hard decision and demodulating equipment utilize the channel response of resulting described another data symbol, and an again data symbol adjacent with another data symbol carried out equilibrium, hard decision and demodulation, and the like.

8. symbol detection apparatus according to claim 6, wherein said frequency pilot sign and described data symbol are the orthogonal frequency division multiplex OFDM symbols that is carried on a plurality of subcarriers.

9. symbol detection apparatus according to claim 6, wherein said mobile communication system are applied to the up link receive diversity of Long Term Evolution LTE system.

Images